Voiceless Spirants Research Articles

Frisian and Norse Sidelights on the Form and Meaning of Proto-Germanic *hlaþanan

Abstract The Germanic verb hlaþanan, usually glossed ‘to load’, has been wrongly evaluated because of a failure to examine the material in sufficient depth as regards both phonology and semantics. The Old High German evidence indicating a regular Verner’s Law alternation in the paradigm is supplemented not only by the Gothic past participle -hlaþans* with analogical voiceless spirant but also by modern East and North Frisian data. This points to a pre-Germanic proto-form with root-final *-t (Part 1). As emerges from a study of the situation in North Germanic, a core meaning for the Germanic verb appears to have been ‘to lay something down (flat)’, from which development to ‘to stack’ and ‘to load’ is straightforward (Part 2). Potential cognates exist in Balto-Slavic, but the exact formal relationships are unclear (Part 3).

Lexica afroasiatica vi

The paper presents the fourth part of a longer series devoted to the discussion of new Afro-Asiatic (Semito-Hamitic) etymologies. The present contribution proposes a set of new lexical roots with *f- in the Anlaut.

Lexica Afroasiatica IX

Comparative-historical Afro-Asiatic linguistics has undergone significant development over the past half century, since the appearance of Essai comparatif sur le vocabulaire et la phonétique du chamitosémitique (1947) by Marcel Cohen. This revolutionary and fundamental synthesis concluded the second great period of the comparative research on Afro-Asiatic lexicon (the so-called “old school”, cf. EDE I 2–4). During the third period (second half of the 20th century), whose beginning was hallmarked by the names of J.H. Greenberg and I.M. Diakonoff, a huge quantity of new lexical material (both descriptive and comparative) has been published, including a few most recent attempts (either unfinishe or rather problematic) at compiling an Afro-Asiatic comparative dictionary (SISAJ a I–III , HCVA I–V, HSED , Ehret 1995).  During my current work on the Etymological Dictionary of Egyptian (EDE ), I have collected a great number of new AA parallels, which – to the best of my knowledge – have not yet been proposed in the literature or were observed independently from me. Along the EDE project (and the underlying “Egyptian etymological word catalogue”), I have started collecting AA roots (not attested in Egyptian) for a separate Afro-Asiatic root catalogue in late 1999.  The series Lexica Afroasiatica started in 2002 for communicating new Afro-Asiatic lexical correspondences observed recently during my work, which may later serve as basis of a new synthesis of the Afro-Asiatic comparative lexicon. The present part of this series is a collection of additional new Afro- Asiatic etymologies with the Proto-Afro-Asiatic initial bilabial nasal (*m-) observed after my research periods at Institut für Afrikanische Sprachwisenschaften of Frankfurt a/M (in 1999–2000 and 2002) guided by Prof. H. Jungraithmayr. The numeration of the etymological entries is continuous beginning from the first part of the series Lexica Afroasiatica.  Each entry is headed by the proposed PAA root (as tentatively reconstructed by myself). Author names are placed after the quoted linguistic forms in square brackets [] mostly in an abbreviated form (a key can be found at the end of the paper). The lexical data in the individual lexicon entries have been arranged in the order of the current classification of the Afro-Asiatic daughter languages (originating from J. H. Greenberg 1955; 1963 and I. M. Diakonoff 1965) in 5 (or 6) equivalent branches: (1) Semitic, (2) Egyptian, (3) Berber, (4) Cushitic, (5) Omotic (cometimes conceived as West Cushitic), (6) Chadic. For a detailed list of all daughter languages cf. EDE I 9–34. The number of vertical strokes indicates the closeness of the language units from which data are quoted: ||| separate branches (the 6 largest units within the family), || groups (such as East vs. South Cushitic or West vs. East Chadic), while | divides data from diverse sub-groups (e.g., Angas-Sura vs. North Bauchi within West Chadic).  Since we know little about the Proto-Afro-Asiatic vowel system, the proposed list of the reconstructed Proto-Afro-Asiatic forms is arranged according to consonantal roots (even the nominal roots). Sometimes, nevertheless, it was possible to establish the root vowel, which is given in the paper additionally in brackets. The lexical parallels suggested herein, are based on the preliminary results in reconstructing the consonant correspondences achieved by the Russian team of I. M. Diakonoff (available in a number of publications) as well as on my own observations refining the Russian results (most importantly Takács 2001). The most important results can be summarized as follows. The labial triad *b-*p-*f remained unchanged in Egyptian, South Cushitic, and Chadic, while the dental series *d-*t-*ṭ was kept as such by Semitic and South Cushitic (AA *ṭ continued as *ḍ in Berber, Cushitic and Chadic, and it was merged into t vs. d in Egyptian). The fine distinction of the diverse sibilant affricates and spirants (AA *c, *ʒ, *2, *s, *č, *ǯ, *2ˇ, *š, *ĉ, *2ˆ, *ŝ) was best preserved in Semitic, South Cushitic and West Chadic (while some of these phonemes suffered a merger in other branches and groups). The Russian scholars assumed a triad of postvelar (uvulear) stops with a voiceless spirant counterpart: *g, *q˙, *q, and *ḫ, the distinction of which was retained in Cushitic and Chadic, but was merged into *ḫ in Semitic and Egyptian. In a number of cases, however, it is still difficult to reconstruct exactly the root consonants on the basis of the available cognates (esp. when these are from the modern branches, e.g., Berber, Cushitic-Omotic, or Chadic). In such cases, the corresponding capitals are used (denoting only the place of articulation). This part contains new Afro-Asiatic roots with *n- followed by a labial. &nbsp

Gothicsiuns, the Domain of Verner's Law and the Relative Chronology of Grimm's, Verner's and Kluge's Laws in Germanic

In various articles published over the last two decades or so, Frederik Kortlandt (1978, 1988, 1991; reaffirmed 1996, 2000), building on his theory that the Proto-Indo-European (PIE) mediae were preglottalized voiced stops (Kortlandt 1978 and more especially 1981, 1985) has claimed, contrary to the traditional view (e.g. Voyles 1992, p. 40; Mottausch 1999, p. 47), (1) that the redistribution of consonantal quality on the basis of accentual context known as Verner's law preceded the rise of voiceless spirants in Proto-Germanic (PGm.), which is often, along with other processes, called Grimm's law, and (2) that the voiced (or lenis) spirants b d g were not a feature of PGm. Since there appears to be little detailed discussion of these last two conclusions of Kortlandt's in the literature – and Bernard Mees (personal communication 30/10/1998, p. 6) has even gone so far as to say that »it is incumbent on [me] to demonstrate that there was no early Verner vis-à-vis Grimm, not [my] critics vice versa« – the intention of the remarks below is to offer a detailed assessment of Kortlandt's arguments.

The Spelling of Proto-Germanic /f/ in Old High German

In the present German orthography, /f/ is represented not only by the letter f (and ph in words of foreign origin) but also by v (e.g. vier, Vater, Frevel, Nerv). The alternation of f and v occurs even in words of the same origin, e.g. fiir-vor, fiillen-voll. This convention is a residue of a spelling tradition more than a millennium old. The practice of denoting the Germanic /f/ by other symbols besides f is found in the oldest German written records known to us. The Latin alphabet, which was employed for writing down old German texts, had only one symbol for both the vowel /u/ and the voiced spirant /v/, which had developed from the semivowel /w/. The forms of the letter in both functions were V in the majuscule, u in the minuscule script. When the minuscule form v was employed, it was merely a graphic variant of u and again served for denoting the vowel as well as the consonant. Likewise in old German manuscripts, this grapheme was used for two phonemes. It again denoted the vowel /u/; but the consonant it represented was the voiceless spirant /f/. Since the Latin symbols f and ph were also adopted, several symbols came to be used for /f/ in Old High German.

Snohomish Chameleon Morphology

1 Field research on which this study is based was conducted during June through August, 1963 and 1964, in connection with the survey of Northwest Indian languages sponsored by the University of Washington. This project was made possible in part by a grant to the University from the National Science Foundation, and in part by other funds allocated by the Graduate School Research Committee. I am grateful to Laurence C. Thompson for advice and guidance in this work. In a more general way I am indebted to Melville Jacobs for training in the anthropological approach to an oral literature-background which has made this field research far more productive and meaningful. The term CHAMELEON is from Hockett, Peiping morphophonemics, Lg. 26.63-85 (1950). See also his rewording of a paragraph from Haas' description of Tunica in Two models of grammatical description, Word 10.210-231 (1954). Examples in this paper are cited in phonemic transcription (without enclosing slants). The phonemes of Snohomish are voiceless spirants /s, 1, s, x, xw, h/; voiceless stops /p, t, c, 6, k, q, kw, qw/; glottalized stops /p, t', I, 6, c, k, , kw, qw, ?/; voiced stops /b, d, j, j, g, gw/; voiced continuants /1, y, w/; vowels /i, u, a, a/. /j/ and /J/ are the voiced counterparts of the affricates /c/ and /c/. Snohomish itself was spoken in the region around Port Gardner Bay, including the southern tip of Whidbey Island, and along the Snohomish River as far upstream as Skykomish. The original dialect area thus corresponds roughly to the southern half of Snohomish County. Today the few remaining speakers live on the Tulalip Indian Reservation across the Snohomish River

The Germanic Consonant Shift and the Theories of J. Fourquet: A Critique

One of the most intriguing attempts in recent years to determine the causes underlying both the Germanic and the High German consonant shifts is that of Professor Jean Fourquet as set forth in his monograph Les mutations consonantiques du germanique (Paris, 1948).1 My references are to the second edition of 1956. His methods are essentially those of the Prague School structuralists and the presentday practitioners of what is usually called diachronic phonemics or phonology. Since most Germanists in this country have been trained in the neogrammarian tradition, Fourquet's procedures and conclusions offer an interesting example of a new approach to an old problem. From the neogrammarian viewpoint the first or purely Germanic consonant shift affected essentially three series of Indo-European stops. The traditional statement is that the IE voiceless stops shifted to voiceless spirants; that the voiced aspirates (presumably via an intermediate stage as voiced spirants) developed into voiced stops, and that the IE voiced stops became in Germanic the corresponding voiceless stops. Fourquet, incidentally, considers the High German mutation to be the result of the continuing influence of the same forces that effected certain changes in the phonetic habits character-

A Note on the Proto-Indo-European Resonants in Germanic

all that distinguishes their environments, we conclude that the words which exhibit such marginal developments were bounded by juncture. Medially there was either no juncture or juncture of a different kind. Not all medial developments from IE voiceless stops were voiceless spirants. If the accent preceded the sound in question, it developed as in marginal position. Since a preceding accent does not by itself explain the development of a voiceless stop (it fails to account for the development initially in the word), and since in the marginal development the effect of juncture seems to have been decisive, we conclude that a post-accentual interior form should also be ascribed to the effect of such a juncture. An initial stressed syllable, then, is assumed to have been bounded by two junctures, an initial unstressed syllable by a preceding juncture and, at most, an interior juncture of less openness; examples: Goth. bahan < */+Pi+xan-/, Goth. fadar < */+fa(-)bAr/. This attempt to explain the facts described in Verner's law has a significant advantage over previous phonetics-centered interpretations. The somewhat arbitrary assignment of priority to either the voiced or the voiceless spirant becomes unnecessary. No temporal order needs to be established, neither form must be derived from the other, and it is without immediate interest whether juncture caused voicing or prevented unvoicing. Whether interior pre-accentual position was marked by weak juncture or by no juncture at all, cannot be decided. The occurrence of interior pre-accentual forms at morpheme boundaries in compounds (OHG me;gi-rahs : sahs) has no weight, in view of German compounds like Breisach (without juncture) vs. Breisgau (with juncture). If internal juncture was combined with preceding accent at the time of the developments described in Verner's law, it seems not impossible to assume such a combination also for the time of the Germanic relocation of accent. The disappearance of internal juncture could then very well lead to a weakening of the syllable boundary, as assumed by Lehmann; the interpretation of umlaut as facilitated by this weakening is very persuasive.

The Personal-Pronoun System of Classical Arabic

points and the resulting other changes are observed. Thus, in English we find one boy, one girl, one man, but two boys, two girls, two men. From this one separates out the partials one, boy, girl, man, two, -s, and replacement of a by e (this example is deliberately put in orthographic terms); -s and the replacement of a by e are then grouped together in a class of morphs, a morpheme. We can now give this morpheme a label, if we wish, such as PLURAL, but we could just as well call it something else; the referential meaning (meta-linguistic or 'real' meaning) does not concern us in the analysis, but there is no need to ignore it in choosing our labels. The Arabic material here analyzed was assembled and checked by Rice in the course of analyzing the Classical Arabic morphemic structure. Trager had long noticed the indications of patterning in the Arabic pronouns, and took the opportunity to suggest and then to elaborate the analysis which is now set forth. One additional theoretical point should be made: we call our partials MORPHS and class them into MORPHEMES. If these entities do not correspond to previous 'definitions' of the terms, we are not at all concerned. We are trying to produce an adequate and useful analysis; the labels we use are conveniently chosen where we find them, and are defined by the use we make of them. 1. The paradigms to be analyzed are the following. (In our transcription,' is the glottal stop, indicates vowel length, and H is the pharyngal voiceless spirant.)

Problems of Armenian Phonology I

This is a continuation of two articles published several years ago in Language (30.197-201 [1954]; 31.4-8 [1955]). In these articles I was concerned above all with a reinterpretation of the interrelationship of the various Armenian reflexes of PIE p t k kw. My conclusions were that it was best to assume that the first result of the Proto-Armenian sound shift of Proto-Indo-European voiceless stops was a set of voiceless spirants, and that some allophones of these spirants later merged with other phonemes, while the principal allophones were affected by a further shift, as a result of which they became voiceless aspirated stops. I interpreted Greek evidence as indicating a rather late date for this second shift (about the beginning of the seventh century A.D.). This interpretation has been challenged by Hans Jensen in an article in Aus der byzantinistischen Arbeit der Deutschen Demokcratischen Republik I (1957), and I now deem it possible that the date of the second shift has to be set considerably earlier than I had assumed. On the other hand, I do not find it necessary to revise my basic theory concerning the results of the first or Proto-Armenian sound shift; only the postulation of ProtoArmenian spirants provides us with a consistent explanation of the seemingly erratic distribution of reflexes of Proto-Indo-European voiceless stops.

The Phonemes of Gothic

It seems fitting, in a number of LANGUAGE dedicated to Leonard Bloomfield, to apply the technique of phonemic analysis to one of the Germanic languages with which he has so long been associated. The present study attempts to analyze the segmental phonemes of Gothic, as far as the written evidence allows us to; and to sketch the changes between Proto-Germanic and Gothic.1. Consonants. Bloomfield shows us how, by using the comparative method, we can reconstruct the phonemes of Proto-Germanic and, as a supplementary detail, obtain a rough phonetic description of these phonemes.2 By applying this method, we arrive at the following fifteen Proto-Germanic consonants: three voiceless stops /p t k/, three voiceless spirants /f b x/, three voiced stopspirants /b d g/, two sibilants /s z/, two liquids /1 r/, and two nasals /m n/. If we assume, as part of the comparative method, that any phonetic change from Proto-Indo-European which is shared by all the Germanic languages is, by definition, also shared by Proto-Germanic, we can even make a few statements about the allophones of four of these phonemes: /n/ = (1) [i] before /g/ and /k/.3 Our oldest Germanic documents--excluding Gothic for the moment--show orthographic n before g and k: ON fingr, OE finger, OHG fingar 'finger'; OE drincan, OHG trinkan 'drink'. We interpret this n as [ij] because the modern dialects which preserve the nasal are unanimous

Relics with 'Broad a' in New England Speech

IN American speech the a phoneme* is universally the regular reflex of ME A in all positions except before r, -lm, -if, nasal plus consonant, voiceless spirants. I shall deal in this article with the occurrences of the a phoneme in positions without apparent consonantal influence, in the New England records of the Linguistic Atlas of the United States and Canada. Members of the a phoneme were recorded by the field-workers in the following words: Saturday, matter, salad, cabbage, sabbath, Matthew, bag, mad. a in Saturday occurs as the only pronunciation of informants in Cheshire, Mass., Ferrisburg, Vt., Burlington, Vt., Topsham, Vt., Lyndon, Vt., Canaan, N.H. Beside the regular trisyllabic form there is a form with two syllables, which is particularly common in phrases like Saturday night, Saturday morning, but is evidently losing ground in many places. We find a distribution common with obsolescent forms: its occurrence in

The Source of the Hittite hi-Conjugation

Hittite verbs fall into two conjugations, one of which shows in the first person singular active an ending -mi and the other an ending -6i; it is customary to speak of the mi-conjugation and the 6i-conjugation. Now Greek verbs are commonly classified as pF-verbs and wverbs. Consequently it naturally occurred to Hrozn?, Sprache der Hethiter 101, that, as the Hittite ending -mi is identical with Gk. -$L, so Hitt. -hi ought to be identical with Gk. -w. The first attempt to carry through this suggestion in detail has just been published by Couvreur, Annuaire de l'Institut de Philologie et d'Histoire Orientales et Slaves 4.551-73 (1936). Since the author is one of those who accept the laryngeal hypothesis,j he must identify the consonant of the ending -hi with an IE laryngeal. He has some doubt as to whether he should see in it the voiceless spirant, which he writes h and I shall write x, or the voiced spirant, which he writes ' and I shall write y. Since x is held to induce a-color of a neighboring vowel, the former hypothesis requires the assumption that IE bher5 comes from *bhero-ax or *bhero-xa; but the Stosston of Lith. velN, etc., is against such a development (cf. Hirt, Handbuch der griechischen Lautund Formenlehre2 468). Consequently he prefers to start with *bhero-y. This, however, causes difficulty; for Couvreur, Hett. h2 140-94, holds that original y yields Hittite 6, as against hh from x, while the first person singular of the hi-conjugation ends in -hhi when a vowel precedes the ending (e.g. da-ah-4i). He tries to get around the difficulty by citing such orthographies as da-a-ah-i in three or four passages (all still unpublished) beside hundreds of occurrences of da-a6-ki,

Notes on the Consonants in the Greek of Asia Minor

The change of the Greek aspirates into the voiceless spirants of the modern language was already beginning to appear in some of the ancient dialects. The intermediate stage in this development is naturally that of affricates,ph, th, kh,becomingpf, tp, kxrespectively, a stage seen in such spellings as μετνλλακχóτα. The evidence of the inscriptions shows that the change was not readily effected in Attic, and the clearest mark of this conservatism is the interchange of aspirates and tenues.